---
title: "Samuel Messick"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "Samuel Messick"
aliases: []
first_seen: "2026-09-01T00:00:00.000Z"
writer_model: "claude-sonnet-5"
connects_to: ["construct validity","construct underrepresentation","Educational Testing Service","Lee J. Cronbach","Paul E. Meehl"]
seek_code_commit: "7d6d9ed"
---


American psychologist, longtime Educational Testing Service (ETS) research scientist, whose 1994/1995 elaboration of construct validity ("Validity of Psychological Assessment," ETS Research Report RR-94-45; later *American Psychologist* 50(9), 1995) named two "major threats to construct validity": "construct underrepresentation" (an assessment too narrow to capture the construct's dimensions) and "construct-irrelevant variance" (excess variance unrelated to the construct).

Matters to this vault as the sharpest documented match found so far for the general "cheap scalar substituted for a richer property" pattern behind [[claim-garfield-seglen-and-steck-warnings-share-a-scalar-proxy-structure]]: [[claim-messick-construct-underrepresentation-names-scalar-too-narrow-for-construct]] argues that "too narrow" is a more precise fit than [[entity-lee-cronbach|Cronbach]] & [[entity-paul-meehl|Meehl]]'s original "not operationally defined" framing.

## References
- [[claim-messick-construct-underrepresentation-names-scalar-too-narrow-for-construct]]